Larger Amounts of Visceral Adipose Tissue in Asian Americans

Authors

  • Yong-Woo Park,

    1. Department of Family Medicine, Sungkyunkwan University School of Medicine, Korea
    2. Obesity Research Center, St. Luke's-Roosevelt Hospital, Institute of Human Nutrition, College of Physicians and Surgeons, Columbia University, New York, New York
    Search for more papers by this author
  • David B. Allison,

    1. Obesity Research Center, St. Luke's-Roosevelt Hospital, Institute of Human Nutrition, College of Physicians and Surgeons, Columbia University, New York, New York
    Search for more papers by this author
  • Steven B. Heymsfield,

    1. Obesity Research Center, St. Luke's-Roosevelt Hospital, Institute of Human Nutrition, College of Physicians and Surgeons, Columbia University, New York, New York
    Search for more papers by this author
  • Dympna Gallagher

    Corresponding author
    1. Obesity Research Center, St. Luke's-Roosevelt Hospital, Institute of Human Nutrition, College of Physicians and Surgeons, Columbia University, New York, New York
    Search for more papers by this author

Obesity Research Center, 1090 Amsterdam Avenue, New York, NY 10025. E-mail: dg108@columbia.edu

Abstract

Objective: Excess visceral adipose tissue (VAT) is recognized as an important risk factor for the development of coronary heart disease and type 2 diabetes. Several studies have reported less VAT in African Americans compared with whites. As little is known about the levels of VAT in Asians, we compared whole-body VAT in Asian Americans with European Americans.

Research Methods and Procedures: VAT was measured using whole-body multislice magnetic resonance imaging in 54 women (18 Asian Americans, 36 European Americans) and 53 men (19 Asian Americans, 34 European Americans) with body mass index (measured in kilograms per square meter) < 30. Data were analyzed by multiple regression modeling.

Results: Asian American women had higher log-transformed VAT compared with European American women (p < 0.05), after adjusting for age and total body fat. There was a significant age by race interaction such that race differences in VAT were most evident over the age of 30 years. No differences in VAT could be detected between Asian American and European American men, even after adjusting for potential covariates, including total adiposity. %Discussion: These data are the first to demonstrate higher amounts of VAT in healthy Asian Americans, a finding that suggests normative VAT values or standards derived from whites may not be applicable to Asians.

Introduction

Excess visceral adipose tissue (VAT) is recognized as an important risk factor in the development of coronary heart disease (1, 2) and type 2 diabetes (3, 4). With the availability of imaging technologies, studies using computed tomography and magnetic resonance imaging (MRI) suggest that the amount of intra-abdominal VAT is highly correlated with adverse outcomes (5, 6).

Several reports indicate race differences in the amount of VAT between African American and white adult women (7, 8, 9, 10, 11), men (10, 11), and youth (12, 13, 14). Little is known about the amount of VAT in Asians. VAT has been reported to be highly correlated with cardiovascular risk factors, coronary artery disease, hypertension, and type 2 diabetes in Chinese (15), Japanese (16, 17), and Japanese American populations (18, 19, 20, 21). Because Asians are shorter, weigh less, and have a higher percentage of body fat compared with African Americans and whites for the same body mass index (BMI) (22), consideration must be given when comparing body compartments across ethnic groups. Although there have been no reports comparing VAT between Asians and whites, Asians seem to have similar VAT areas, despite lower subcutaneous adipose tissue areas (16, 18, 19, 20, 23, 24), when compared with studies of VAT areas by single-slice computed tomography among whites (7, 8, 9, 10, 11). We therefore hypothesize that after first controlling for total body fatness, Asians have greater amounts of VAT compared with whites.

The aim of this study was to evaluate the amount of VAT in Asian Americans compared with that in European Americans without and with adjustment for total fatness.

Research Methods and Procedures

Subjects

The subjects were 54 premenopausal women (18 Asian Americans and 36 European Americans) and 53 men (19 Asian Americans and 34 European Americans), ages 18 to 44 years, with a BMI (measured in kilograms per square meter) < 30 who had participated in an ongoing multiethnic body composition investigation (25). Recruitment of subjects occurred over a 3-year period through advertisements in local newspapers, on radio stations, and on flyers posted in the local community. Inclusion criteria required that subjects be ambulatory with no medical conditions that could potentially affect any of the variables under investigation. Each subject completed a medical examination that included screening blood tests. Only healthy subjects, without any diagnosed medical conditions, were enrolled in the study. Race was determined by self-report and all parents and grandparents were required to be Asian and non-Hispanic white for the Asian American and European American subjects, respectively. The Institutional Review Board of St. Luke's–Roosevelt Hospital approved the study, and all subjects gave written consent to participate.

Anthropometry

Body weight was measured to the nearest 0.1 kg (Weight Tronix, New York, NY) and height to the nearest 0.5 cm using a stadiometer (Holtain, Wales, UK). Waist circumference was obtained with a measuring tape at the level of the narrowest part of the torso, as viewed anteriorly. Hip circumference was measured at the level of maximum extension of the buttocks as viewed from the side.

DXA

Total body fat was measured with a whole-body DXA scanner (DPX, Lunar Radiation Corp., Madison, WI) using software version 3.6. Repeated daily measurements in three adult subjects showed a coefficient of variation (CV) of 3.4% for whole-body fat. Ethanol and water bottles (8-liter volume), simulating fat and fat-free soft tissues, respectively, were scanned as soft-tissue quality control markers monthly. The range in measured R values over the study period was 1.255 to 1.258 (CV = 0.127%) and 1.367 to 1.371 (CV = 0.103%), for ethanol and water, respectively.

MRI

Total adipose tissue (TAT), including visceral and subcutaneous adipose tissue volumes, were measured using whole-body multislice MRI. Subjects were placed on the 1.5-T scanner (General Electric, 6X Horizon, Milwaukee, WI) platform with their arms extended above their heads. The protocol involved the acquisition of ∼40 axial images, 10-mm thickness, and at 40-mm intervals across the whole body (26, 27). Images were analyzed using VECT image analysis software (Martel, Montreal, CA) on a Sun Workstation (Silicon Graphics, Mountain View, CA). MRI–volume estimates were converted to mass using the assumed density of 0.92 kg/liter for adipose tissue (28). The technical error for repeated measurements of the same scan by the same observer of MRI-derived TAT and VAT volumes in our laboratory are 1.1 ± 1.2% and 1.1 ± 1.5%, respectively (27).

Statistical Analysis

Gender-specific means and standard deviations for descriptive statistics and body composition variables were compared between race groups. Multiple linear regression models were performed to adjust for age, gender, race, and total body fat. A natural log transformation of VAT was performed to achieve normality and homoscedasticity of residuals. Each gender group was analyzed separately because of a gender by race interaction. Analysis of covariance was used to test for differences in the VAT values after adjusting for covariates between Asian Americans and European Americans. The level of significance was set at p < 0.05 (two-tailed). Statistical calculations were performed using SPSS for Windows version 9.0 (SPSS Inc, Chicago, IL).

Results

Baseline Characteristics

The baseline subject characteristics are summarized in Table 1. As a group, the Asian American women were shorter (p = 0.013) and weighed less (p = 0.006) compared with the European American women. The difference in BMI was marginally significant (p = 0.052) between the two groups of women. There were no significant between-group differences in percentage of body fat (p = 0.911), waist circumference (p = 0.121), and waist-to-hip ratio (p = 0.137).

Table 1.  Baseline characteristics
 Asian AmericansEuropean Americansp value
  • Values are means ± SD; p values are for Asian Americans vs. European Americans.

  • *

    Measured by DXA.

Women (n)1836 
 Age (y)28.6 ± 5.232.8 ± 7.20.030
 Body weight (kg)53.5 ± 7.460.6 ± 9.10.006
 Height (cm)160.5 ± 6.2165.4 ± 6.60.013
 Body mass index (kg/m2)20.7 ± 2.022.1 ± 2.60.052
 Percentage of body fat (%)*25.9 ± 6.925.7 ± 6.60.911
 Waist (cm)67.7 ± 5.470.6 ± 6.90.121
 Hip (cm)89.7 ± 6.195.7 ± 7.40.005
 Waist-hip ratio0.75 ± 0.030.74 ± 0.040.137
Men (n)1934 
 Age (y)29.3 ± 6.932.2 ± 7.00.146
 Body weight (kg)67.1 ± 8.879.7 ± 9.7<0.001
 Height (cm)173.2 ± 5.5179.6 ± 6.40.001
 Body mass index (kg/m2)22.3 ± 2.724.7 ± 2.50.002
 Percentage of body fat (%)*18.0 ± 4.918.0 ± 7.20.996
 Waist (cm)77.8 ± 6.684.9 ± 7.40.001
 Hip (cm)93.5 ± 6.099.5 ± 5.0<0.001
 Waist-hip ratio0.83 ± 0.030.85 ± 0.050.090

Asian American men were shorter (p = 0.001) and weighed less (p < 0.001) compared with the European American men. Asian American men had a lower BMI (p = 0.002), smaller waist circumference (p < 0.001), and lower waist-to-hip ratio (p = 0.090) compared with the European American men. There was no significant difference between the two groups of men in percentage of body fat (p = 0.996).

VAT Mass and Distribution

There were no significant differences between the two groups of women in total (p = 0.111), visceral (p = 0.121), and subcutaneous (p = 0.175) adipose tissue mass (Table 2). The relative proportion of TAT as VAT was significantly higher in Asian American compared with European American women (p = 0.006).

Table 2.  Body composition results
 Asian AmericansEuropean Americansp value
  • Values are mean ± SD; p values are for Asian vs. European Americans.

  • *

    Visceral adipose tissue (percentage of TAT): (visceral adipose tissue/total adipose tissue) × 100.

Women (n)1836 
 Total adipose tissue (kg)14.2 ± 4.616.5 ± 5.20.111
 Visceral adipose tissue (kg)0.74 ± 0.460.56 ± 0.340.121
 Subcutaneous adipose tissue (kg)13.4 ± 4.315.9 ± 5.00.175
 Visceral adipose tissue (percentage of TAT)*5.1 ± 2.73.4 ± 1.60.006
Men (n)1934 
 Total adipose tissue (kg)12.1 ± 3.015.5 ± 5.80.022
 Visceral adipose tissue (kg)1.1 ± 0.61.7 ± 1.10.043
 Subcutaneous adipose tissue (kg)10.8 ± 2.713.7 ± 5.00.027
 Visceral adipose tissue (percentage of TAT)*9.0 ± 4.210.1 ± 4.70.372

Asian American men had less total (p = 0.022), visceral (p = 0.043), and subcutaneous (p = 0.027) adipose tissue mass compared with the European American men. There was no significant difference in the relative proportion of TAT as VAT (p = 0.372) between the two groups of men.

Regression Models

There was no significant between-group difference in log-transformed VAT without adjustment for covariates observed in the women (p = 0.207).

After adjusting for age, there was a significant (p = 0.033) racial difference in log VAT between Asian American and European American women. The significant difference remained unchanged after adding total body fat (p = 0.025) (Table 3). A significant interaction between age and race was found in the women (Figure 1). The Johnson-Neyman technique (29) was therefore used to find the point where the two regression lines intersected. The intersection point was 23.7 years, and the 95% confidence region for the intersection was 10.4 to 29.5 years. This indicates that for women above 29.5 years, the Asian American women have significantly more VAT compared with European American women. With increasing age, the racial difference in log VAT increased among women over 30 years. However, below that age, a significant difference between the two groups was not observed.

Table 3.  Multiple regression analysis between log VAT and adjusted variables in women
 Dependent variable = log VAT
Independent variablesRegression coefficient (SE)p valuer2SEE
  • *

    Dummy codes are 0, European American and 1, Asian American.

  • Interaction variable between age and race.

Intercept−1.169 (0.441)0.011  
Age−0.006 (0.012)0.607  
Race*−1.897 (0.820)0.02537.8%0.48
Age× race0.077 (0.028)0.008  
Total body fat0.047 (0.012)<0.001  
Figure 1.

Association between log VAT vs. age in Asian American (•) and European American (○) women. Oblique solid and dashed lines represent the simple linear regression model for Asian Americans (y = −2.265 + 0.065x, r = 0.50, p = 0.04) and European Americans (y = −0.845 + 0.0005x, r = 0.07, p = 0.71), respectively. There was a significant intersection between the two lines.

There was no significant racial difference in log VAT between Asian American and European American men (p = 0.101). No significant between-group differences in VAT were observed, even after adjusting for potential covariates.

Statistical adjustment of the mean log-transformed VAT values by analysis of covariance between Asian Americans and European Americans are shown in Table 4. After adjusting for age and body fatness (BMI or TAT), Asian American women had higher mean values of log VAT compared with the European American women. There was no significant difference in VAT in the men with or without adjustment.

Table 4.  Estimated log VAT by ANCOVA
 Estimated log VAT (SE)  
CovariatesAsian AmericansEuropean AmericansF test*p value
  • Evaluated covariates appeared in the model: in women, age, 31.6 yr, BMI = 21.6 kg/m2, TAT = 15.7 kg; in men, age, 31.0 yr, BMI = 23.7 kg/m2, TAT = 13.3 kg.

  • *

    F tests are the effect of race variable.

  • In women, age × race (0, European American, 1, Asian American) interaction variable is included.

Women (n)1836  
 No covariate−0.42 (0.14)−0.63 (0.10)1.630.207
 Age−0.22 (0.15)−0.62 (0.10)4.970.030
 Age, BMI−0.17 (0.15)−0.65 (0.09)7.560.008
 Age, TAT−0.16 (0.13)−0.67 (0.08)10.930.002
Men (n)1934  
 No covariate0.04 (0.16)0.37 (0.12)2.790.101
 Age0.13 (0.15)0.32 (0.11)1.050.311
 Age, BMI0.20 (0.10)0.36 (0.14)0.730.398
 Age, TAT0.23 (0.08)0.30 (0.11)0.280.601

Discussion

The present study results partially support the hypothesis that Asian Americans have more VAT than European Americans, after controlling first for potential covariates. The hypothesis proved correct in the women but VAT was not significantly different between the two groups of men. The VAT difference in the women was most evident in subjects over the age of 30 years.

A precedent exists for race differences in VAT and upper body obesity along with concomitant metabolic abnormalities. Unwin et al. (30) compared glucose tolerance between Chinese and European adults living in the United Kingdom. The prevalence of glucose intolerance in the Chinese subjects was similar to or higher than that in Europeans, despite a significantly lower mean BMI and waist circumference in the Chinese subjects. A relatively higher VAT in Asians compared with whites may be one explanation for these less favorable metabolic abnormalities.

With respect to other race groups, Dowling and Pi-Sunyer (31) reported that upper body obesity is less detrimental with respect to health risks for African Americans compared with white women of similar total adiposity. Conway et al. (7) found significantly less VAT in African Americans compared with white women. African Americans had significantly lower levels of plasma glucose and triglycerides, with significantly higher levels of high-density lipoprotein cholesterol as compared with white women. Hill et al. (9) reported that for any given body fatness, African American men had significantly less VAT than white men. The differences in VAT between the two groups of women were not significantly different when corrected for total body fatness. Després et al. (10) found that African American women had levels of VAT similar to those of white women although African American women had higher body fat amount than did white women. A lower accumulation of VAT in African Americans is accompanied by significantly fewer metabolic abnormalities compared with whites. Thus, there may be broad race differences in the inter-relations between VAT, metabolic abnormalities, and health risks.

It is estimated that the current prevalence rates of type 2 diabetes in Asian countries exceeds 8% of the adult population (32). This rapid increase is attributed to reductions in physical activity through increased use of automobiles and an increase in dietary fat, a trend synonymous with that reported in many Western countries (33). Fujimoto et al. (34) reported that Japanese Americans had a higher prevalence of type 2 diabetes than Japanese living in Japan, which suggests that Westernization plays a role. In addition to environmental factors, genetic susceptibility to visceral fat accumulation may accelerate the increased prevalence of type 2 diabetes. Further research is needed to confirm genetic and environmental involvement.

Body surface measurements such as waist circumference and waist-hip ratio are commonly used to predict visceral fat accumulation in epidemiological studies. The current study results suggest that the cut-off points of these surrogates for visceral obesity may be race-specific.

This study measured the amount of total VAT by whole-body multislice MRI. MRI has the advantage of producing high-quality images without exposing the subject to radiation; therefore, repeated scanning or multiple-slice scanning is possible. Movement of intra-abdominal contents, either from breathing or bowel motion, may increase the measurement error for intra-abdominal fat. This error can be decreased through the acquisition of multiple slices rather than a single slice (35). Another advantage of multislice abdominal imaging over a single-slice is that total VAT as a percent of total body adipose tissue can be calculated. Similar to percentage of body fat being used clinically as a diagnostic criterion and follow-up measurement, percentage of TAT as VAT may be more meaningful as an index of visceral obesity. The Asian American women in our study had a significantly higher percentage of TAT as VAT compared with their European American counterparts, even without adjustment for other covariates.

With respect to study limitations, subjects were volunteers recruited as a convenience sample. In the current study, Asian American subjects were persons residing in the United States and no attempt was made to control for years living in the United States for Asian-born subjects. Lifestyle factors such as smoking, alcohol ingestion, and physical activity level were not considered in the analyses, which could be confounders (36). Because Asian Americans with a BMI ≥ 30 kg/m2 are uncommon in the general population compared with heavier European Americans, the current sample was restricted to subjects with a BMI < 30 kg/m2. Our subject pool was also relatively young (age ∼30 years), and thus, the extrapolation of these findings to elderly and obese people may be inappropriate.

In conclusion, these data are the first to demonstrate higher amounts of VAT in healthy, non-obese Asian American women. These findings suggest that normative values or standards for VAT derived from whites may not be applicable to Asians. Additional studies of larger and more age- and weight-diverse subject populations could clarify some of the issues raised in the present report.

Acknowledgments

Supported in part by National Institutes of Health Grants P01-DK42618, R01DK51716, and P30DK26687, and an educational grant from Bristol Myers Squibb/Mead Johnson.

Ancillary